This course introduces students to regression as a tool to answer questions about education. Regression is commonly used to answer questions about (A) “association claims” – about the relationship between variables – and (B) “causal claims” – about the causal effect of one variable on another. However, using regression appropriately requires thoughtfulness about what kinds of questions regression can answer, about the assumptions regression relies on, about the limitations of our data, and about how particular variables (e.g., “race” and “gender”) are incorporated into analyses. Otherwise, regression results may be biased and may reify rather than interrogate problematic ideas. Therefore, EDUC152 introduces students to fundamental concepts of regression analysis and how these concepts can be thoughtfully applied to address important questions about education. The course also trains students how to read and critically assess research that uses regression. ECUC152 integrates theory and application using the R programming language. Students will be assessed through four substantive take-home assignments, including the final, capstone assignment.
[HAVE NOT UPDATED SINCE LAST MEETING]
Regression is the most widely-used quantitative methodology to answer causal, and also non-causal, research questions. This section of EDUC152 will introduce students to regression with a focus on using regression to answer causal research questions, which typicaly follow the form “what is the effect of X on Y.” The course also emphasizes undersanding how to read and critically assess empirical research that uses regression.
The course integrates statistical theory and application using the R programming language. Students will work through asynchronous video lectures and lectures slides on their own. These lectures introduces statistical theory, introduces the relevant programming skills, and provides the code and real-world data so that students can practice conducting and interpreting statistical analyses. Course topics will include: fundamental statistical concepts of statistical inference; principles of causal inference; and fundamentals of multiple regression. During class time, students will work in groups to solve practical research challenges and we will discuss and deconstruct empirical research that uses regression analysis. The primary course assessments are four problem sets – including the final capstone problem set – which will be completed in groups. Each problem set will require students to apply knowledge of statistical concepts, and conduct substantive statistical analyses around a particular research question.
The course embraces using regression to answer traditional research questions (e.g., the effect of student-teacher ratio on achievement) and critical research questions (e.g., the effect of racial salience – as presented in email text – on how white university admissions officers respond to inquiries from Black prospective students). The skills this course teaches are valued by employers and are valued in the process of applying to graduate schools. After completing this course, students will be prepared to take more advanced causal inference coursework (e.g., quasi-experimental methods) and coursework that teaches the programming and data manipulation skills necessary to create analysis datasets for real research projects.
Big-picture (conceptual) learning goals
COOMMENTS
Skill-based learning goals
Each week, the course will be structured around asynchronous (pre-class) lectures and one synchronous workshop-style class meeting per week. Weekly homework will consist of students working through the lectures on their own and a modest amount of required reading. Written homework will consist of four “problem sets.” Students will complete the first three problem sets in groups. Students will complete the final capstone problem set, due during finals week, on their own.
Prior to our in-class meetings, students should work through lecture materials on their own. We recommend treating the lecture materials as an active learning experience, in which students run R code on their computer instead of merely reading text on the slide. Additionally, we recommend that students ask questions on the course github website when they are having difficulty with the material.
With respect to written work, the problem sets – described below – will be substantive and are intended to be challenging. Students who devote time each week working through the lecture materials will be better prepared for the problem sets. We recommend starting the problem sets early. This way students will have plenty of time to ask for help on questions they find challenging.
We all have a responsibility to ensure that every member of the class feels valued and safe. Be mindful that our words and body language affects others in ways we not fully understand. We have a responsibility to express our ideas in a way that doesn’t make disparaging generalizations and doesn’t make people feel excluded. As an instructor, I am responsible for setting an example through my own conduct.
Learning regression, while trying to get a handle on R and unfamiliar data can feel overwhelming! We must create an environment where students feel comfortable asking questions and talking about what they did not understand. Discomfort is part of the learning process. Unburdern yourself from the weight of being an “expert.” Focus your energy on improving and helping your classmates improve.
Every course should be an anti-racist course, even when the subject matter is broadly oriented. In this course we’ll engage with research that reflect systemic gaps based on race, ethnicity, immigration status, and gender identity, among other aspects of identity. We will discuss whether the language, the framing, the analyses, and even the research question of a study contribute to problematic beliefs. If so, how can we do better? It is also critical that we acknowledge that the social and economic marginalization reflected in data is rooted in systemic oppression that upholds opportunity for some at the expense of others. We should all be thinking about our own role in upholding these systems.
COMMENT
All course related material can be found on the class website [here]. Pre-recorded lecture videos, lecture slides (pdf, html), and .Rmd files will be posted on the class website under the associated sections. Additional resources (e.g. syllabus) may also be posted on the class website. Class announcements and discussion will be conducted on GitHub (see below).
COMMENTS:
COMMITTED TO USING SLACK INSTEAD OF GITHUB
GET LANGUAGE FROM GLORY ON SLACK IN SYLLABUS
if you are going to use github for course communication; then this should be tied to a learning goal; `
We will be using GitHub teams for class announcements.
GitHub teams: The teaching team will post all class announcements using GitHub teams. The GitHub team discussions feature allows for quick and seamless communication to all members of an organization or team– in this case, to all students with a GitHub account enrolled in the course. Some features include:
@mentioned by all students enrolled in the class and part of the organization.Credit: Introducing team discussions
We will be using GitHub issues for questions and class discussion.
GitHub issues: GitHub issues are traditionally used by collaborators of a repository for managing tasks for a project. Our rational for using issues is twofold: 1) help track and organize questions related to course material and problem sets and 2) promote classroom participation. Students are encouraged to contribute to issues by posting questions, sharing helpful resources, and/or taking a stab at answering questions posted on issues. Some features include:
If you have a personal question or issue, you can email the instructor or TA directly. Additionally, we are available for office hours or by appointment if there is anything you would like to discuss with us in private.
Required books
Optional books
Links to other required and optional reading will be on the course website [link]
Required software
Course grade will be based on the following components:
The primary course assessments are four problem sets. The first three problem sets are each worth 15% of the course grade and students will work in groups. The final, capstone problem set is worth 30% and students will work alone.
Each problem set will require students to apply knowledge of statistical concepts, conduct substantive statistical analyses, present and interpret results. Other questions will introduce students to some of the thorny data challenges that inevitably arise in real research projects. The capstone problem set will require students to conduct the major components of an empirical regression analysis, from research question and variable collection to modeling, presentation, and interpretation. Additionally, the capstone problem set will require students to critically evaluate an empirical journal article that utilized the same data sources to answer the same research question.
You will work in groups for the first three problem sets. However, it is important that you understand how to do the problem set on your own, rather than copying the solution developed by group members. Each student will submit their own R script or .Rmd file. Since you will be working together, it is understandable that answers for some questions will be the same as your group members. However, if I find compelling evidence that a student merely copied solutions from a classmate, I will consider this a violation of academic integrity and that student will receive a zero for the homework assignment.
Late submissions will lose 20% (i.e., max grade becomes 80%). Problem sets not submitted by XXXX will not receive points. You will not lose points for late submission if you cannot submit a problem set due to an unexpected emergency. But please contact the instructor by email as soon as you can so we can work out a plan.
We strongly recommend using GitHub issues to ask questions you have about problem sets. Instructors will do our best to reply quickly with helpful hints/explanations and we encourage members of the class to do the same.
More detailed problem set guidelines can be found here [link]
COMMENT:
During most synchronous class sessions, students will work on a group activity or challenge that applies concepts and skills we are learning. For example, Run a statistical test in R and use the test results object to create a formatted table that you insert into a .Rmd file. Or, write a draft critique of the methodology of a published research paper. Students will submit their work at the end of class. Most of these tasks cannot be fully completed. The goal is to get students thinking. Students will be graded largely based on effort.
Students are expected to participate in the weekly class meetings by being attentive, supportive, by asking questions, or by answering questions posed by others on zoom. Additionally, students can receive strong participation grades by asking questions and answering questions on GitHub issues.
| Letter Grade | Percentage |
|---|---|
| A+ | 99-100% |
| A | 93<99% |
| A- | 90<93% |
| B+ | 87<90% |
| B | 83<87% |
| B- | 80<83% |
| C+ | 77<80% |
| C | 73<77% |
| C- | 70<73% |
| D | 60<70% |
| F | 0<60% |
UNITS:
[HAVE NOT UPDATED SINCE LAST MEETING]
FUNDAMENTALS OF REGRESSION IN DESCRIPTIVE 1. Introduction to bivariate regression 1. Prediction and measures of model fit 1. Hypothesis testing and confidence intervals for \(\beta_1\) 1. Categorical X variables and introduction to multivariate regression 1. INSERT A WEEK ON CRITICALLY ASSESSING RESEARCH; AND HOW TO AVOID DOING BAD THINGS PEOPLE DO WITH REGRESSION REGRESSION FOR CAUSAL INFERENCE 1. Principles of causal inference: Rubin’s Causal Model and why experiments work 1. Using regression for causal inference: OLS assumptions and omitted variable bias
COMMENTS - ON UNITS, FROM GLORY - “The kinds of quant research claims we discuss in ED 150 are descriptive claims, association claims, causal claims. ED 150 alum will find it helpful - I think - if you’re able to reference these (especially the second two) when you talk about internal validity and transition from your unit on”regression as a descriptive tool" to your unit on “regression as a causal tool.”" - WANTING TO SEE EXAMPLES OF EMPIRICAL READING; - REPLACE “WHY EXPERIMENTS WORK” WITH SOMETHING ELSE; 1. Creating tables and graphs of descriptive statistics and regression results x. when to come up with FOR CDAS, NEED TO EXPAND THIS OUT
Center for Accessible Education
Students needing academic accommodations based on a disability should contact the Center for Accessible Education (CAE). When possible, students should contact the CAE within the first two weeks of the term as reasonable notice is needed to coordinate accommodations. For more information visit www.cae.ucla.edu.
Located in A255 Murphy Hall: (310) 825-1501, TDD (310) 206-6083; http://www.cae.ucla.edu/
UCLA policy
This class
As a student you may experience a range of issues that can cause barriers to learning, such as strained relationships, increased anxiety, alcohol/drug problems, depression, difficulty concentrating and/or lack of motivation. These mental health concerns or stressful events may lead to diminished academic performance or reduce a student’s ability to participate in daily activities. UC offers services to assist you with addressing these and other concerns you may be experiencing. If you or someone you know are suffering from any of the aforementioned conditions, consider utilizing the confidential mental health services available on campus.
Students in distress may speak directly with a counselor 24/7 at (310) 825-0768, or may call 911; located in Wooden Center West; www.caps.ucla.edu
UCLA is committed to maintaining a campus community that provides the stronget possible support for the intellectual and personal growth of all its members- students, faculty, and staff. Acts intended to create a hostile climate are unacceptable.
The LGBTQ resource center provides a range of education and advocacy services supporting intersectional identity development. It fosters unity; wellness; and an open, safe, inclusive environment for lesbian, gay, bisexual, intersex, transgender, queer, asexual, questioning, and same-gender-loving students, their families, and the entire campus community. Find it in the Student Activities Center, or via email lgbt@lgbt.ucla.edu.
The Dashew Center provides a range of programs to promote cross-cultural learning, language improvement, and cultural adjustment. Their programs include trips in the LA area, performances, and on-campus events and workshops.
This program provides a safe space for undergraduate and graduate undocument students. USP supports the UndocuBruin community through personalized services and resources, programs, and workshops.
UCLA Student Legal Services provides a range of legal support to all registered and enrolled UCLA students. Some of their services include:
Due to COVID, Student legal Services is closed to walk-ins.
UCLA Students with Dependents provides support to UCLA studens who are parents, guardians, and caregivers. Some of their services include:
For more information visit their website: https://www.swd.ucla.edu/
Lactation Rooms
Gender Inclusive restrooms
Campus accessibility
Title IX prohibits gender discrimination, including sexual harassment, domestic and dating violence, sexual assault, and stalking. If you have experienced sexual harassment or sexual violence, there are a variety of resources to assist you.
CONFIDENTIAL RESOURCES:You can receive confidential support and advocacy at the CARE Advocacy Office for Sexual and Gender-Based Violence, A233 Murphy Hall, CAREadvocate@careprogram.ucla.edu, (310) 206-2465. Counseling and Psychological Services (CAPS) also provides confidential counseling to all students and can be reached 24/7 at (310) 825-0768.
NON-CONFIDENTIAL RESOURCES: You can also report sexual violence or sexual harassment directly to the University’s Title IX Coordinator, 2255 Murphy Hall, titleix@conet.ucla.edu, (310) 206-3417. Reports to law enforcement can be made to UCPD at (310) 825-1491. These offices may be required to pursue an official investigation.
Faculty and TAs are required under the UC Policy on Sexual Violence and Sexual Harassment to inform the Title IX Coordinator should they become aware that you or any other student has experienced sexual violence or sexual harassment.
Peer learning facilitators (PLFs) are undergraduates who understand the challenges of writing at UCLA. Scheduled appointment and walk-in options are available, see www.wp.ucla.edu/uwc for more information about writing programs and to get assistance with your writing.
For additional campus resources and student services, please review this document.